I feel like we're saying basically the same thing, but I feel like I want to elaborate. I'm gonna start off by saying that I'm an engineering student, so I'm pretty well versed when it comes to math models. Your way of assessing draft picks works (pretty well) when trying to calculate a general value of a draft pick given its position, but it lacks A LOT when looking at specific examples.
For example: if the model concludes that a first overall pick is worth X, say a first line forward for 10 years, something like a Mitch Marner or a Leon Draisaitl. This would be great when looking at the overall value of a first overall pick (I do feel like you're undervaluing a first overall pick a bit when you say that Brock Boeser is performing like one, but nevertheless), and predicting future first overall pick and their values. But if you'd apply this model to a draft in the past, say the 2015 draft you'd run into huge problems.
We know that the first overall pick in 2015 was Connor McDavid, a generational talent who at this stage of his career looks like he's well on his way to becoming at least a top 10 player of all time. If you applied your model to this draft it would say that Peter Chiarelli would've had a great draft had he chosen Jack Eichel with the first overall pick (as Eichel is outperforming what a first overall pick has historically done), and that he'd gotten a passing grade had he chosen Marner, as Mitch looks to be about what you can expect from a first overall pick (given this model).
I think that you'd have a hard time finding a single fan who would agree that the Oilers would've had a good draft, had they passed on McDavid and chosen either Eichel or Marner in 2015.
If you look at the 2012 draft on the other hand, I would argue that the first real impact player picked was Filip Forsberg at #11, and I wouldn't call him a "first overall level talent". The 2012 draft was historically weak and leading in to the draft the consensus was that either Yakupov or Ryan Murray would go first. Saying that the Oilers had a bad draft when they picked Yakupov (who turned out being one of the worst first overall picks of the 21st century) is true, but saying that they should've picked Filip Forsberg is unrealistic, since no GM in the league would have done that given the Oiler's position, and even so, even given that Forsberg is the best player in the entire draft, he's still worse than most first overall picks are, thus the model would tell us that even he was chosen first overall (again, really unlikely since he wasn't considered a first overall talet at the time) whoever picked him would have had a bad draft.
My point in this is that when assessing such a small number of drafts as we are when talking about Benning, you have to look at them one by one to get a good idea of how good a job he's done, this being because the sample size is too small and that the expected value of a draft pick oscillates so much from year to year. The thing with mathematical models based on probabilities and stats work great when used with a really large sample size. you can, given enough data tell the difference between 50.01 % and 49.99 %, but given a small sample size, it can be impossible to tell 1/3 from 2/3. Again, what I'm saying is that the model (although sophisticated and based in facts) is not a good tool when trying to do what you're doing: assess single drafts.
I genuinely think that you have to look at Benning's drafts one by one and look at the picks that were made around his picks to tell how good a job he's done. And when doing this, I think it's clear to say that he's missed by quiet a wide margin on some of the picks (mainly in the 2014 and 2016 drafts), hit home runs in some (2015 and 2017), and that the 2018 draft is looking really good at the moment. And it is true that all GMs miss in the draft, but people on here go on and on about what a "draft wizard" he is, and I can't help but feel that if you've missed on two first rounders (both picked in the very beginning of their drafts) in a five year span; maybe you can't be considered a draft wizard.